During the time that the vessel was secured at the container terminal working cargo, pre-heat was maintained on the engine. Jacket water was circulated through the engine. Lubricating oil was continuously circulated to the main and crosshead bearings and fuel oil was continuously circulated through the fuel oil system. Cylinder liner lubrication is not in operation during shut-down periods, as oil flow to the hydraulic motor driving the cylinder liner lubricating oil pumps is regulated according to engine fuel demand. Under emergency running-conditions, oil from the cross-head bearing oil system may be used to drive the cylinder-lubricating pump by opening and adjusting a valve which by-passes the flow regulating valve. The by-pass valve was reportedly found to be leaking slightly during the post-explosion examination of the engine by the vessel's engineers. It was also reported that the cylinder-lubrication oil storage tank contained about 230 litres less than it should have when checked after the explosion. With the by-pass valve leaking and the engine stopped, oil would have been supplied to the cylinder lubricators. Depending upon the position of the piston within the cylinder, oil could have accumulated on the piston crowns. Prior to starting the engine, the normal procedure is to open the indicator cock on each cylinder head and turn the engine over with the turning gear. An accumulation of fuel, oil or water in the cylinders would be indicated by a spray from the indicator cock as the engine was turned over. There were no reports of any liquid being sprayed out of the indicator cocks. This evidence indicates that there was no significant accumulation of any liquid in any of the cylinders prior to starting the engine. With the by-pass valve open and the engine running, excessive cylinder-liner lubrication oil would have been supplied to the engine during the voyage from Europe to Canada. Oil vapour carry-over into the exhaust system would have occurred. Given that there was no damage nor indication of a fire in the exhaust trunk, no overspeed of or damage to the turbochargers, the slight bulging and tearing of the lower hopper and no damage to any of the tubes in the economizer, then the main force of the explosion appears to have taken the path of least resistance, travelling upwards, and dislodging the rain hood on its way. The physical evidence and the damage observed would indicate that the explosion was initiated at or near the top of the exhaust gas heat exchanger. Furthermore, this area was reportedly clear of the light soot that was in evidence throughout the rest of the exhaust system. The lack of light soot suggests that this area had been subjected to the most intense heat and therefore was the seat of the explosion. It was not possible to positively determine the source of the inflammable vapours. Given the loss of 230 litres of cylinder liner oil, and given that the injectors were reported to be in satisfactory condition, then it is possible that some of the cylinder liner oil carried over into the economizer while the engine was running. Ignition was caused, in all likelihood, by a spark carried in the exhaust gas stream when the engine was started.Analysis During the time that the vessel was secured at the container terminal working cargo, pre-heat was maintained on the engine. Jacket water was circulated through the engine. Lubricating oil was continuously circulated to the main and crosshead bearings and fuel oil was continuously circulated through the fuel oil system. Cylinder liner lubrication is not in operation during shut-down periods, as oil flow to the hydraulic motor driving the cylinder liner lubricating oil pumps is regulated according to engine fuel demand. Under emergency running-conditions, oil from the cross-head bearing oil system may be used to drive the cylinder-lubricating pump by opening and adjusting a valve which by-passes the flow regulating valve. The by-pass valve was reportedly found to be leaking slightly during the post-explosion examination of the engine by the vessel's engineers. It was also reported that the cylinder-lubrication oil storage tank contained about 230 litres less than it should have when checked after the explosion. With the by-pass valve leaking and the engine stopped, oil would have been supplied to the cylinder lubricators. Depending upon the position of the piston within the cylinder, oil could have accumulated on the piston crowns. Prior to starting the engine, the normal procedure is to open the indicator cock on each cylinder head and turn the engine over with the turning gear. An accumulation of fuel, oil or water in the cylinders would be indicated by a spray from the indicator cock as the engine was turned over. There were no reports of any liquid being sprayed out of the indicator cocks. This evidence indicates that there was no significant accumulation of any liquid in any of the cylinders prior to starting the engine. With the by-pass valve open and the engine running, excessive cylinder-liner lubrication oil would have been supplied to the engine during the voyage from Europe to Canada. Oil vapour carry-over into the exhaust system would have occurred. Given that there was no damage nor indication of a fire in the exhaust trunk, no overspeed of or damage to the turbochargers, the slight bulging and tearing of the lower hopper and no damage to any of the tubes in the economizer, then the main force of the explosion appears to have taken the path of least resistance, travelling upwards, and dislodging the rain hood on its way. The physical evidence and the damage observed would indicate that the explosion was initiated at or near the top of the exhaust gas heat exchanger. Furthermore, this area was reportedly clear of the light soot that was in evidence throughout the rest of the exhaust system. The lack of light soot suggests that this area had been subjected to the most intense heat and therefore was the seat of the explosion. It was not possible to positively determine the source of the inflammable vapours. Given the loss of 230 litres of cylinder liner oil, and given that the injectors were reported to be in satisfactory condition, then it is possible that some of the cylinder liner oil carried over into the economizer while the engine was running. Ignition was caused, in all likelihood, by a spark carried in the exhaust gas stream when the engine was started. An explosion occurred in the upper section of the economizer in the main engine exhaust gas system when flammable vapours in the upper section of the economizer ignited after the engine was started. The force of the explosion was sufficient to eject the rain hood from the stack top and cause damage to the lower hopper. It was not possible to determine the type of combustible material present in the exhaust system which ignited.Findings An explosion occurred in the upper section of the economizer in the main engine exhaust gas system when flammable vapours in the upper section of the economizer ignited after the engine was started. The force of the explosion was sufficient to eject the rain hood from the stack top and cause damage to the lower hopper. It was not possible to determine the type of combustible material present in the exhaust system which ignited. The explosion in the exhaust system of the containership CAPECHARLES resulted from the ignition of flammable gases in the upper section of the economizer.Causes and Contributing Factors The explosion in the exhaust system of the containership CAPECHARLES resulted from the ignition of flammable gases in the upper section of the economizer.